Engine
Package engine is Hanzo Engine: which models the serving runtime has loaded, and the GPUs under it.
Package engine is Hanzo Engine: which models the serving runtime has loaded, and the GPUs under it.
| Base URL | https://api.hanzo.ai |
| Operations | 4 |
| Auth | Authorization: Bearer $HANZO_API_KEY |
engine
GET /v1/engine/model
Read one model's load state on the serving runtime
Model reads one model's load state — loaded, unloading, or not_found, as the engine itself reports it.
| Parameter | In | Type | Required | Description |
|---|---|---|---|---|
model | query | string | — | Model is the model id to inspect, exactly as the model list reports it. |
GET /v1/engine/models
List the models the serving runtime holds, with each one's load state
Models lists the models the engine serves, each with its load state — the server's own model table (its standard list envelope, load status included), relayed verbatim.
GET /v1/engine/status
Whether the serving runtime is reachable, and which build it runs
Status reports whether the engine deployment is reachable and which build revision it runs — an honest lens for "is the serving runtime up", never a fabricated ok.
GET /v1/engine/system
The serving host's own inventory: devices, memory and build capabilities
System reads the engine host's inventory: OS, CPU, memory, every accelerator device with its VRAM and compute capability, and the build's capabilities (CUDA/Metal/flash-attention) — the real hardware under the serving runtime, relayed verbatim.
How is this guide?